Two thirds of patients lost to follow up after four years of fee for service treatment

Many of the first patients to receive care in sub-Saharan Africa had to pay for it out of their own pocket. Those that could afford to stay on an ART regimen (and who survived long enough to stay on treatment), responded quite well according to a long-term study of a fee-for-service programme in Uganda, published in the April issue of the Journal of Acquired Immune Deficiency Syndrome.

Patients in the study who remained on treatment had good long-term (up to at least four years) viral load and CD4 cell responses. However, many patients in the study were unable to continue paying for continuous treatment on their own.

Fee-for-service programmes

“The demand by patients who pay for drugs in the private sector or at nongovernmental hospitals has in many ways led the way for access to drugs,” wrote Kabugo et al. Before antiretroviral drug prices came down and made national public health programmes offering ART possible, some of the first pilot treatment programmes in Africa sought to simply monitor the safety and effectiveness of ART in patients paying for treatment.

This study monitored virologic and immunologic responses in patients receiving treatment at the St. Francis Hospital, Nsambya, a non-governmental tertiary care referral hospital located in Kampala, Uganda. Although participants had to pay for their own antiretroviral drugs, viral load and CD4 cell monitoring were provided for free.

The study

Three hundred twenty-one patients in Uganda attended the HIV clinic between 1998 and 2002. At study entry, 34% were antiretroviral experienced, the median CD4 count was 79 cells, and the median viral load was 249,489 copies/ml.

Two hundred sixty-three (82%) patients returned at least once after the initial visit; of these, 54 (21%) had to interrupt treatment (mainly because they couldn’t afford it) for over a year. When the price of therapy began to come down in 2000, due to the availability of generic ART, many patients returned to the study.

By 2002, one hundred thirty-five patients were still in care. 69 patients were known to have died (only 9 of whom died in 2002), and 68 were lost to follow-up. The probability of patients surviving and remaining in care at one year was only 0.56 (95% confidence interval [CI]: 0.50–0.61), falling to 0.35 (95% CI: 0.29–0.41) at 4 years.

But these fortunate patients did well on treatment. After the first year, the median CD4 cell count increase was 55 cells, by year two, it was 112 cells, 142 cells at year three, and 131 cells at year 4. The median log viral load decrease from baseline after 1 year was 1.4 log10 copies/mL, 1.32 log10 copies/mL after year 2, 1.9 log10copies/mL at year 3, and 1.51 log10 copies/mL by year 4.

Complicating factors

On-treatment analyses usually overstate the benefit of the therapy analysed since poor responders to treatment generally go off drug and their negative data are not Included In the final analysis. However, the bias can be forgiven in this instance since the purpose of this study was to show that good long-term benefits can be derived in this challenging setting. As the authors wrote “it is encouraging to find that thosewho manage to stay on therapy can derive demonstrable longterm benefits in Africa in a manner that would be expected in industrialized countries.”

Part of the reason it had been decided to run such a fee-for service clinic was to provide an alternative to donor-supported programmes that might not be sustainable. But the fee-for-service context presented researchers with unique challenges.

Since patients were paying for treatment, their choices (for reasons of cost or perception) determined the regimen selected. For example, the preferred nucleoside analog backbone combination changed when one combination became cheaper.

Another feature of fee-for-service programmes is that, in contrast to public sector programmes, there is little reliance on the use of counselors and other support staff because “the funds to hire support staff would need to be recouped through higher prices for services or drugs.” And due to cultural or privacy concerns, “some patients in a fee-for-service setting may not want to meet with staff other than the doctor.” As a result doctors either had to spend a great deal of time with many patients or patients suffered by not receiving enough attention.

However, as long as parallel free-ART programmes exist for those who cannot afford to access fee-for-service programmes, the researchers of this study still support continuing the fee-for-service model that primarily pays for itself — because it guarantees the continuation of a functioning HIV clinic that patients can return to when and if donor-supported clinics fail.

To help retain patients, and to reduce the burden on those accessing treatment from the private sector will require “adequate access to trained providers, stabilization of costs for drugs and laboratory monitoring at a level that is manageable to clients, as well as an increase in the support services available. We need not wait until all foreseeable obstacles have been overcome; …with adequate resources, we can continue to implement and expand effective programs and bring treatment to people living in the countries with the worst HIV epidemics in the world.”